404 REPORT OF COMMISSIONER OF FISH AND FISHERIES. 



Family NOTACANTHID-ai. 



29. Macdonaldia challengeri (Vaillant). 



XotavaiithiiH rinsuanus (Hiuther, CLallenj^er Report, vol. xxil, p. 250, pi. LXi, 

 lig. B; not of Filippi and Veraui. 



Vaillant was perfectly Justified in separating this Pacific form from the Mediter- 

 ranean X. rixsudtiKu, with which Giinther had identified it. The lower, heavierspines 

 in both dorsal and anal fins, the more anterior orij^in of the dorsal, which is a little in 

 advance of base of pectorals, the very short robust ventral spine, and the lower 

 insertion of the pectoral fin sufficiently distinguish the species, in addition to the 

 jjeculiarities in the shape of the snout and the greatly increased number of anal 

 spines, to which Vaillant calls attention. 



The Albatross dredged a single specimen, 500 mm. long, at station 3308, west of 

 Pribilof Islands in Bering Sea, at a depth of 1,625 fathoms. Giinther's description, 

 above cited, of a fish taken south of Yeddo at a depth of 1,875 fathoms, agrees so well 

 with our specimen that no doubt can exist of their identity. The maxillary spine, 

 not shown in Giinther's figure, is very evident in our specimen. The brancliiostegal 

 rays are distinctly 6 instead of 5 in number, and the caudal contains 5 instead of 6 

 rays. There are 35 dorsal spines. The anal spines pass so gradually into the rays 

 that they are distinguishable with difficulty. Definite articulations ajtpear before 

 the rays have lost their spinous character, while still stiff and pungent. Dividing 

 them on the basis of these articulations, the anal fin contains 27 spines and about 

 153 soft rays. 



Family GASTEROSTEIDiE. The Sticklebacks. 



30. Pygosteus pungitius (Linnseus). 



Several specimens were secured from the vicinity of Nushagak, one from the 

 Naknek River, and another from the nest of a sea bird on Round Island, of the Wal- 

 rus Island group, all in Bristol Bay. None of our specimens shows the short ventral 

 spines ascribed to P. pungitius brachypoda, their length being in every case 2^ to 2| 

 length of head. Brachypoda was originally described by Dr. Bean from Greenland, 

 and has been given in his recent lists as the common form of Alaska. Awaiting 

 further information, we refer our specimens rather to typical pungitius. 



31. Gasterosteus cataphractus (Pallas). 



Abundant at Departux'e Bay, Vancouver Island, May 10-13. 



Family AMMODYTIDiE. 



32. Ammodytes personatus Girard. Sand Lance. 



Unalaska, Chernofi'ski, Ilerendeen Bay, Hagemeister Island, and generally in 

 shallow water. It forms an important element in the food of the codfish. 



Family BERYCIDiE. 



33. Melamphaes lugubris Gilbert. 



One specimen from north of Unalaska, station 3327, depth 322 fathoms. 



Family BATHYMASTERIDiE. 



34. Bathymaster signatus Cope. 



Taken very abundantly inour series of shallow-water dredgings along the southern 

 shore of the Alaskan Peninsula, northward through Unimak Pass and north of 

 Unalaska. The stations at which it was obtained are numbered 3211, 3212, 3213, 

 3214, 3215, 3217, 3220, 3222, 3223, and 3319 ; the depths range from 34 to 56 fathoms. 

 In addition, a very few small specimens were secured at stations 3262, 3309, 3321, and 



